CONCEPTO DE PUESTO DE TRABAJO

The importance of self-reflexivity in qualitative research is evident in interviews, since the interviewer is also a participant in the research. As such, interviews offer both risks and opportunities to the researcher. The risk of researcher bias accidentally shaping the results versus the opportunity of data that offers insights and knowledge not previously known.

A Semi Structured Interview Approach.

Although the interview protocol (Appendix 9B) appears on paper to be structured, the actual approach to the interviews was semi-structured in that points raised by the participants in answers were immediately followed up and the participants were also given opportunities to raise their own points or add their own commentary on items

not on the interview protocol, so that the questions were open-ended. In this way the protocol was used as a prompt for the interviewer but mostly dialogue flowed

between the interviewer and participant in a formal but responsive way. The

interview protocol lists the main questions and some possible follow-up questions but probes were also used as questions during the interview, when possible, and it was these probes and follow-up questions that demonstrated the responsivity in the interviews, according to Rubin & Rubin’s description of responsive interviewing (Rubin & Rubin, 1995:16). The operational difficulties of conducting international interviews over the phone should be highlighted here too, as well as the difficulties in gaining access through gatekeepers, gaining the confidence of participants and

establishing a rapport when there is no face to face validation (Creswell, 2007: 120). It was also extremely difficult to retain access and participation. With it taking months to secure both participants and interviews in many cases and almost a year in the case of the TNC participant it was also a challenge to secure a date for the

interview slot with the participant or gatekeeper, even when permission had been granted in principle. With perseverance it was possible to conduct semi-structured interviews, to audiotape and then transcribe them, which constituted the data collection. The one part of the interview process which proceeded easily, however, was the issuance and signage of ethics approval sheets and participant information sheets.

As a lone method the interviews would not have been robust since not enough

interviewees chose to participate, so they are not representative of each sector but are representative of the GFG field in a holistic sense and their perspectives on the global governance of food are represented. As a supplementary method, however, the interviews offer useful new insights into the food governance field. The interviews combined with the case studies offer an endogenous view of the GFG field that augments the Field Analysis. This endogenous view of the GFG field is important to develop issues raised in the literature review such as power-sharing between actors,

silo mentality and the effects of corporations, to collect recommendations for improvement such as inclusion and scrutiny mechanisms and also to appraise in a wider sense how food security is understood amongst the agency, commercial and charity workers who work on it.

Research Process and Participant Selection

As mentioned there was significant difficulty in securing participants to take part in the interviews and the challenge was two-fold in that it was also difficult to get them to reply to contact, or participate in the interview. In some cases it took persistent emailing, verbal coaxing and assurances to obtain participation. One food security expert explained that this was typical and the issue was suspicion across the food security governance field. Mostly Aid organisations replied, sometimes with offer of participation, probably because they have a public facing role and remit, with a mission statement to raise awareness and they also seek funding from grant-awarding sources. 90 potential respondents were approached with a request for an interview from a master list of 107 organisations (see Appendix 3). The master list devised for the Field Analysis shows that the personnel are a mixture of senior management and field workers. The head of the organisation was approached first and it was often then delegated by them to a more junior (and perhaps more relevant) person. For more generic organisations, research was conducted into the most relevant role to approach. The most typical response from many was either non-response or declining to get involved. Another common response was to agree but then to not respond to the interview phone call.

A total of ten respondents were interviewed from nine organisations; seven by phone, two face-to-face (in the same interview) and one completed the interview protocol by email. An additional two replied to a later follow up email interview although this was a truncated version of the questions (see Appendix 9 for versions of the protocol). The responses to this email largely tended to point out website links.

The interviews are anonymised and each of the four sectors are represented by at least one interview. The ‘aid’ sector (AID) offered six interviews (from five organisations, participants B, C, D, E and G) and the other three sectors have at least one participant each so there is one interview each with a national Governmental Organisation (Participant A), an International Organisation (Participant F), a Trans National

Corporation (Participant H) and a Civil Society Organisation (Participant I) (see Table 6 below). The Governmental Organisation interview has been used here although not part of the quadripartite sectors of the GFG field. One of the Aid Organisations was the pilot interview. The gender balance is six men and four women. Three

participants are based in the UK, three in Europe and four in the USA and there was a wide span of the working age demographic.41 It is regrettable that Asia and Africa are not represented here. This perhaps reflects the geographical bias of the field, the constraints of unfunded doctoral research and the domain of the researcher.

A GO

B Aid (pilot interview) C Aid D Aid E1 E2 Aid F IO G Aid H TNC I CSO

Table 6: Interview Participants’ code by Sector

Although, as mentioned, each sector is represented, the interviews are not fully representative of the total GFG field. The findings from these two methods

41 _{Although demographic detail was not formally asked for, this data was ascertainable from research}

(interviews and case studies) are illustrative and, combined with the Field Analysis, provide a foundation for observations, analysis and some recommendations. Due to the low participation rate in the interviews there are, however, a number of ‘missing voices’. As well as other geographical areas, the biggest player in each sector, namely Monsanto, the Gates Foundation, World Food Programme and La Via Campesina all declined to be interviewed. As a result these organisations are chosen as the case studies so that their profile could be set out. Future research into the GFG field that both overcomes this Euro-American bias and specifically includes

interviews with the larger players would offer a further important step forward in understanding Food Security. Unfortunately this was outwith the reach of this research.

Although the difficulties in securing participants were consistent and enduring, no pattern can be drawn from non-participation in the interview other than it was a common response, given that more than 80 other participants also declined. In the most extreme example it took twelve months to obtain an interview, that of the TNC. It was very important to the research so polite persistence and perseverance was imperative. In the same way that abstaining from voting is a form of democratic expression, not participating in the interview can be seen as an expression towards the question, the investigation and perhaps also towards academic research generally. Email correspondence with a number of potential participants drew positive

comments on the importance and relevance on the research and frequently there were requests for a briefing paper or some form of dissemination of the results but there can only be speculation on the reasons for their subsequent non-involvement. Referral to websites or corporate publications was another typical response, which shows a preference for unidirectional mode of communication. Some, such as La Via Campesina, have a policy of non-disclosure via online, by publication or even by interview, but overall these non-disclosing organisations tended to be non US-based organisations. Interestingly and of note for future researchers, La Via Campesina’s position is to reject academicism (and so studies are rare).

Of the nine organisations who did participate, eight gave at least 30 minutes of interview time and were generous in offering supplementary information. Advice sought from five food security or global governance experts confirmed that these responses of non-participation are a typical stance and sectors only deal with or care

about their own sector so pan sectorial co-operation and oversight are rare. This may go some way to explain why an overall specification of the membership of the field has been absent. The interviews were conducted from August -September 2013 with the preparation starting in February 2013. The results do indicate differing points of view from across sectors and show contradicting and competing opinions and understandings of effective GFG.

The Interviewer’s Approach

As mentioned, although the interview protocol appears to be structured (see Appendix 9) the approach for the interviews was intentionally more semi-structured. Open- ended and open questions were asked and answers which were interesting or opened up new topics were followed up with further off-schedule questions as part of the semi-structured approach. The interviewees were also asked if they wanted to raise any part of global food governance, or comment on any issues. This enabled

narratives to be established and the responsiveness gave the space for a wider range of observational and analytical data to be drawn from the participants (Appendix 10). Reflections on the Pilot Interview and the Questions

Reframing the interview questions and nomenclature awareness became apparent during the conduct of the pilot so post-pilot, a second version of the interview

protocol was devised (Appendix 9, B). The pilot/first version had 32 questions which were a mix of primary and secondary questions and duplication, overlap and some slight leading of the participant were evident but this initial version was a necessary precursor to creating the final version of the protocol (see Appendix 9, A). The main reflections from the pilot are on the praxis of the research, (such as the recording equipment failing to record and so the importance of detailed note taking as a backup). The semi structured approach taken in conducting the survey was again

demonstrated by asking the participants for comment, or any other points they wanted to add.

For the next interviews, the participants’ publications were read and ‘brief sheets’ started on their organisation, work etc. In the pilot I slightly dominated the interview, not letting the voice of the participant come through well enough but I did stay objective, only steering the participant to remain ‘on topic’, not drifting off onto seed or advocacy (which the interviewee wanted to). When piloting the questions I also realised that there was still a slight mix of primary and secondary questions together under topics such as reform, field, and regulation so this ordering was sorted for the remaining interviews. Plus the pilot interview took too long – overall the

interviewees gave much more comprehensive questions than expected and also anticipated questions, so to get around interviewee fatigue I rearranged the order of some of the questions, frontloading the most arduous questions that required the highest cognitive load to the beginning. I also collapsed the first two questions on personal role/organisational involvement since most interviewees spoke only about their organisational involvement with a brief sentence on their role involvement. I also frequently asked if they had issues, points or questions. Overall it went well and operationalising the interviews brought the topic to life and the enabled the data collection.

Interview Themes

Once redrafted, the final version of the interview protocol (Appendix 9, B) worked well. The questions for the interview were divided into three sections of personal involvement, GFG and reform of GFG, each with an open question on that theme at the end. The conclusion was one open question on any aspect of the topic. There were thirteen questions in all which covered the participant’s role in food security; asked for their evaluation of the effectiveness of current GFG; the sectors in GFG; the architecture, oversight, reform including possible leadership of any reform; scrutiny of TNCs, Aid organisations and International Organisations and also opinions on profood security policy innovations.

Towards the end of the interview, if time allowed, a question was asked about which issues the G8 should prioritise (although this question was also implicitly asking what

it was going to take to improve food security). Answers given to question four (on the main sectors of the GFG field) were applied to the Field Analysis. A point on nomenclature was raised by a few interviewees at the start, since food security and global governance definitions do vary but this was quickly reconciled during the interview.42

Logistical Difficulties

For all of the post-pilot interviews physical challenges predominated. For example the logistics of recording over poor phone signals, time zone alignment, unkept appointments, note-taking at speed, interruptions and delays were all repeatedly overcome. But the biggest challenge remained the lack of uptake, sometimes on the day. The response rate really depended on the head of the organisation. The focus of the interviews was to capture opinions rather than facts (since facts are in the public domain) and the questions were designed to avoid duplication with publicly available information. Once the interviews were all concluded an email ‘sweep’ of unfulfilled interviewees was conducted with a truncated version of the interview protocol of five questions. In research terms I felt that by the time I had completed the interviews, my decision to do the Field Analysis was again validated in that there was no map of GFG nor overall sense of actors involved even amongst many of the governors

themselves. The lack of interview uptake and tentativeness of the interviewees meant that interviews alone would not yield a complete view of who was involved in GFG since the actors and sectors themselves are very siloed and Balkanized.

42 _{Interestingly another observation is that the nomenclature of ‘Non Governmental Organisation’ did}

not transfer across sectors. The danger is of non-fully articulating definitions is misspecification but a few participants pointed to the need for aligned nomenclature as a prequel to their interviews.

The only face-to-face interview possible was with one aid organisation and this was a much better quality of data-yield than the rest. Face-to-face interviews as standard would have been better since the data quality was so much improved. So another observation is that it is not easy to do organisational interviews on a very low budget and with no prior relationship nor access. Despite being persistent and the importance of the topic there was almost no response from the ‘Roman Forum’, International Organisations and TNCs. The faith based organisations, however, did reply. This widespread lack of reply is a reflexive conclusion in itself but also speaks to the opaqueness of the field. Perhaps Bourdieu’s concept of social capital also applies to social researchers?

Paradoxically, for a governance field so static, another of the challenges of the

research generally was incorporating the numerous developments in such a vast topic area. For example, between starting and completing the research, the Gates

Foundation came out from the ‘left field’ so the Field Analysis was revised to include the Gates Foundation and was changed from three sectors to four sectors. Measuring the effects of the Gates Foundation on the GFG field could be an entire research project in itself. Discovering that LVC did not disclose data any more was another surprising development. With their secretariat now based in Zimbabwe and a recent policy of non-transparency (with the justification of not revealing funders or funding) LVC’s non-disclosure reset the narrative of the research into a more complex, shaded picture. These developing changes justify the methodology because by consistently asking systematic questions across sectors and not accepting only the received wisdom, interesting and sometimes surprising results can be obtained.

To sum up, despite thorough planning, carrying out the research was an exercise in solving a series of problems. Obtaining ‘thick, rich data’ was difficult but since the purpose was to garner endogenous viewpoints from the global governance field, enough new ‘thick, rich’ data was obtained. The response rate was not ideal but this also unburdened this researcher from one of a researcher’s more judicious decisions. Rather than stopping when the data was satiated, the interviews ended when all the respondents who were willing to get involved were interviewed.

The Interview Data

The interview themes in the protocol are drawn from the themes arising from Chapter Three namely, governance architecture, dynamics, effectiveness, power and

membership. Taking the first major theme, membership, the first question asked the participant about the role and involvement of their organisation in food security. The Aid Organisations commented that their organisations either fed into UN

organisations e.g. the Committee on Food Security; promoted agricultural

development (Participant C); conducted campaigns, lobbied and offered advocacy services (Participants D and G), or pursued economic-based approaches to food insecurity. This shows the wide range of activities by the Aid Organisations who participated in the interviews. The Governmental Organisation provides oversight and scrutiny of the food security activities of the legislative arm of the US

Government. The International Organisation commented that it achieves economic development through agricultural growth. The TNC stated that it improves the supply chain by supporting farmers to access markets. Lastly Participant I, the CSO,

commented on their role in GFG:

“[We] work with the CFS, not as a member (only states are members), but as an active participant in the Civil Society Mechanism (CSM) to the CFS. [Organisation name] has, in the context of the respective CSM working group, co-facilitated civil society input to two important processes in the CFS: the Voluntary Guidelines on Responsible Governance of Tenure of Land, Forests and Fisheries, adopted by the CFS in May 2012, and to the Global Strategic Framework for Food Security and Nutrition, adopted by the GSF in October 2012.”

These answers to the first question provide descriptive rather than analytical data but are useful in identifying how the actors view their involvement in food security. But it is clear that for some of the respondents’ secondary activities, such as feeding into